Big data and cloud are possibly the two most hyped concepts in IT today. And while both have their place in the CIO’s toolbox of solutions, neither is the kind of silver bullet they sometimes are touted to be. We have looked at the possibilities offered by big data previously in this blog, so let’s now have a look at what cloud as a deployment model has to offer BSS.
There are generally two key arguments used by vendors to support the idea of deploying BSS in the cloud as opposed to on premises. The first one of these is cost – or rather a pay-as-you-grow pricing model – and the other one is flexibility. From the point of view of the CSP, these are in reality the same thing; the traditional wisdom of the price of flexibility still holds true no matter what the deployment model, but at the same time lack of flexibility at the wrong time can be priceless.
Deploying BSS in the cloud, however, brings to the table new considerations as well. The first of these is reliability: Deploying business-critical pieces of your infrastructure at a remote location surely has its drawbacks. Regulatory issues have often been stated as a major stumbling block here, but in reality these are reasonably easily overcome in real life. A more relevant problem, however, may well be setting up reliable connectivity. Not so much a problem in the West, but a very relevant concern in Africa, for example, where otherwise redundant routes regularly converge at critical junctions of the internet, thus presenting a single point of failure.
Very often then, we see CSPs starting to look at solutions where the real-time aspects such as charging are deployed locally, with back-office functions centralized in private or public clouds, thus offering a best-of-both-worlds approach. But why stop there? BSS is actually a collection of reasonably independent, loosely coupled functions, each of which can be deployed where and how the best value for money can be extracted. For example, many of the back-office functions where little telecom specific knowledge is required can be easily outsourced to companies performing these functions to a wide range of industries – often even in the CSP's home country, thus avoiding any extended regulatory battles.